Data Visualization with R — Base Graphics

Aravind Hebbali

References

Everything this book depends on, in one place. The primary sources for base graphics are the R manuals that ship with the software — they are free, they are current, and ?graphics is more accurate than any blog post.

Base R graphics

R Core Team (2026). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Available at https://www.R-project.org/.

Chambers, J. M. (1993). Statistical Models in S. Chapman & Hall, London.

Chambers, J. M. (2008). Software for Data Analysis: R and S-PLUS. Springer, New York.

Chambers, J. M. (2018). Scripting R: Analyzing and Programming with the R Language. O’Reilly Media, Sebastopol, CA.

Murrell, P. (2017). R Graphics, 3rd edition. Chapman & Hall/CRC, Boca Raton, FL. The definitive reference for base graphics specifically. ?graphics and the R Graphics help pages are abridged versions of it.

Venables, W. N. and Ripley, B. D. (2002). Modern Applied Statistics with S. Springer, New York. The S ancestor of base graphics, and still the clearest account of why the design is shaped the way it is.

The grammar of graphics, and the alternative

Wilkinson, L. (2005). The Grammar of Graphics. Springer, New York.

Wickham, H. and Grolemund, G. (2017). R for Data Science: Import, Tidy, Transform, Visualize, and Model Data. O’Reilly Media, Sebastopol, CA. Available at https://r4ds.hadley.nz/.

Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis, 3rd edition. Springer, New York.

Chang, W. (2024). R Graphics Cookbook, 2nd edition. O’Reilly Media, Sebastopol, CA. Available at https://r-graphics.org/. Recipe-first, and covers base and ggplot2 side by side.

Colour and accessibility

Okabe, M. and Ito, K. (2008). “Color Universal Design (CUD): How to make figures and presentations that are friendly to Colorblind people.” In Proceedings of the 2008 Symposium on Universal Design and Accessibility, Tokyo, Japan.

Masataka, D. (2020). Color Design: Methods, Tools, and Applications. Apress, Singapore. Background on perceptual colour spaces, lightness and contrast that Chapter 12’s greyscale check relies on.

The Okabe-Ito palette used in Chapter 12 is available in R 4.1 and later without any additional package, via palette.colors(palette = 'Okabe-Ito').

Datasets used in this book

Most figures are drawn from datasets that ship with R, and the two that do not are vendored in this repository so the book builds offline. Full provenance:

AirPassengers — Box & Jenkins (1975) monthly airline passengers, 1949–1960. Distributed with R from the datasets package.
mtcars — extracted from the 1974 Motor Trend US magazine. Distributed with R from the datasets package.
iris — Anderson, E. (1935). “The irises of the Gaspé Peninsula.” Bulletin of the American Iris Society, 59, 17–50. Distributed with R from the datasets package.
ToothGrowth — the growth of 60 guinea pigs, distributed with R from the datasets package.
hsb2.csv — High School and Beyond survey, 200 observations, originally from the UCLA Institute for Digital Research and Education. Vendored as data/hsb2.csv. Used illustratively; not redistributable as an authoritative source, and no claim is made about the sampling frame.
brics-gdp-2010-14.csv — a frozen vintage of annual GDP growth for the BRICS nations, 2010–2014, hand-entered as a teaching dataset. It differs slightly from the World Bank’s current revised series and should be treated as illustrative, not as a citable statistic. Where the numbers matter, use the World Bank directly.

Reproducing the measurements in Chapter 1

Chapter 1’s claims are measured, not asserted. Each figure in that chapter is drawn from a CSV committed to this repository, and each CSV is produced by a script in scripts/. None of them runs when the book is built, which is deliberate: a book arguing for zero dependencies cannot quietly require two of them to make its case.

Claim	Script	Output
Draw time and peak memory, base versus ggplot2, 10k and 100k points	`scripts/bench.R`	`data/bench-base-vs-ggplot2.csv`
Wall clock to start a plotting session, by engine	`scripts/startup.R`	`data/startup-latency.csv`
Declared non-base CRAN dependencies, per package	`scripts/cran-deps.R`	`data/cran-deps.csv`
Published container image sizes, R versus R + tidyverse	`scripts/docker-sizes.R`	`data/docker-sizes.csv`

Run any of them with Rscript scripts/<name>.R. Each stamps the date, R version and platform it ran under into its output, so a stale row is visibly stale. Absolute timings are machine-specific and should be read as an order of magnitude; the shape of each result travels across machines far better than the individual numbers do.

scripts/bench.R requires bench and ggplot2, which the book itself never needs. scripts/docker-sizes.R and scripts/cran-deps.R read published metadata over HTTP, so their numbers reflect the state of the world on their retrieved_on date.

A note on the sources for this chapter

The author packages above were checked by reading their source on 2026-10-04, not inferred from dependency metadata. Dependency metadata records what a package needs, not what it draws with, and drawing that distinction incorrectly is exactly the error Chapter 1 warns about.

--- description: "Sources cited by Data Visualization with R - Base Graphics: the R manuals and papers that define base graphics, the colour vision research behind the Okabe-Ito palette, and provenance for every dataset and benchmark in the book." --- # References {.unnumbered} Everything this book depends on, in one place. The primary sources for base graphics are the R manuals that ship with the software — they are free, they are current, and `?graphics` is more accurate than any blog post. ## Base R graphics > R Core Team (2026). *R: A Language and Environment for Statistical Computing*. > R Foundation for Statistical Computing, Vienna, Austria. > Available at <https://www.R-project.org/>. > Chambers, J. M. (1993). *Statistical Models in S*. Chapman & Hall, London. > Chambers, J. M. (2008). *Software for Data Analysis: R and S-PLUS*. Springer, New York. > Chambers, J. M. (2018). *Scripting R: Analyzing and Programming with the R Language*. O'Reilly Media, Sebastopol, CA. > Murrell, P. (2017). *R Graphics*, 3rd edition. Chapman & Hall/CRC, Boca Raton, FL. > The definitive reference for base graphics specifically. `?graphics` and the > `R Graphics` help pages are abridged versions of it. > Venables, W. N. and Ripley, B. D. (2002). *Modern Applied Statistics with S*. > Springer, New York. The `S` ancestor of base graphics, and still the clearest > account of why the design is shaped the way it is. ## The grammar of graphics, and the alternative > Wilkinson, L. (2005). *The Grammar of Graphics*. Springer, New York. > Wickham, H. and Grolemund, G. (2017). *R for Data Science: Import, Tidy, > Transform, Visualize, and Model Data*. O'Reilly Media, Sebastopol, CA. > Available at <https://r4ds.hadley.nz/>. > Wickham, H. (2016). *ggplot2: Elegant Graphics for Data Analysis*, 3rd > edition. Springer, New York. > Chang, W. (2024). *R Graphics Cookbook*, 2nd edition. O'Reilly Media, > Sebastopol, CA. Available at <https://r-graphics.org/>. > Recipe-first, and covers base and ggplot2 side by side. ## Colour and accessibility > Okabe, M. and Ito, K. (2008). "Color Universal Design (CUD): How to make > figures and presentations that are friendly to Colorblind people." In > *Proceedings of the 2008 Symposium on Universal Design and Accessibility*, > Tokyo, Japan. > Masataka, D. (2020). *Color Design: Methods, Tools, and Applications*. > Apress, Singapore. Background on perceptual colour spaces, lightness and > contrast that Chapter 12's greyscale check relies on. The Okabe-Ito palette used in Chapter 12 is available in R 4.1 and later without any additional package, via `palette.colors(palette = 'Okabe-Ito')`. ## Datasets used in this book Most figures are drawn from datasets that ship with R, and the two that do not are vendored in this repository so the book builds offline. Full provenance: - `AirPassengers` — Box & Jenkins (1975) monthly airline passengers, 1949–1960. Distributed with R from the `datasets` package. - `mtcars` — extracted from the 1974 *Motor Trend* US magazine. Distributed with R from the `datasets` package. - `iris` — Anderson, E. (1935). "The irises of the Gaspé Peninsula." *Bulletin of the American Iris Society*, 59, 17–50. Distributed with R from the `datasets` package. - `ToothGrowth` — the growth of 60 guinea pigs, distributed with R from the `datasets` package. - `hsb2.csv` — High School and Beyond survey, 200 observations, originally from the UCLA Institute for Digital Research and Education. Vendored as `data/hsb2.csv`. Used illustratively; not redistributable as an authoritative source, and no claim is made about the sampling frame. - `brics-gdp-2010-14.csv` — a frozen vintage of annual GDP growth for the BRICS nations, 2010–2014, hand-entered as a teaching dataset. It differs slightly from the World Bank's current revised series and should be treated as illustrative, **not** as a citable statistic. Where the numbers matter, use the World Bank directly. ## Reproducing the measurements in Chapter 1 Chapter 1's claims are measured, not asserted. Each figure in that chapter is drawn from a CSV committed to this repository, and each CSV is produced by a script in `scripts/`. None of them runs when the book is built, which is deliberate: a book arguing for zero dependencies cannot quietly require two of them to make its case. | Claim | Script | Output | |:--|:--|:--| | Draw time and peak memory, base versus ggplot2, 10k and 100k points | `scripts/bench.R` | `data/bench-base-vs-ggplot2.csv` | | Wall clock to start a plotting session, by engine | `scripts/startup.R` | `data/startup-latency.csv` | | Declared non-base CRAN dependencies, per package | `scripts/cran-deps.R` | `data/cran-deps.csv` | | Published container image sizes, R versus R + tidyverse | `scripts/docker-sizes.R` | `data/docker-sizes.csv` | Run any of them with `Rscript scripts/<name>.R`. Each stamps the date, R version and platform it ran under into its output, so a stale row is visibly stale. Absolute timings are machine-specific and should be read as an order of magnitude; the shape of each result travels across machines far better than the individual numbers do. `scripts/bench.R` requires `bench` and `ggplot2`, which the book itself never needs. `scripts/docker-sizes.R` and `scripts/cran-deps.R` read published metadata over HTTP, so their numbers reflect the state of the world on their `retrieved_on` date. ## Further reading in the R ecosystem The author's own CRAN packages are adjacent to this book's subject matter rather than part of it. Their plotting layers use ggplot2; Chapter 1 discusses that honestly, and Appendix A maps the two systems onto each other. - [`olsrr`](https://github.com/rsquaredacademy/olsrr) — companion functions for ordinary least squares, including diagnostic plots. - [`rfm`](https://github.com/rsquaredacademy/rfm) — recency, frequency and monetary analysis, with cohort and segment plots. - [`blorr`](https://github.com/rsquaredacademy/blorr) — logistic regression diagnostics and plots. - [`descriptr`](https://github.com/rsquaredacademy/descriptr) — frequency tables, cross-tabulations and summary statistics, each with a plot. ## A note on the sources for this chapter The author packages above were checked by reading their source on **2026-10-04**, not inferred from dependency metadata. Dependency metadata records what a package *needs*, not what it *draws with*, and drawing that distinction incorrectly is exactly the error Chapter 1 warns about.